Granted Patent B2

US 12,430,816 · App. 17/807,644 · Granted Sep 30, 2025

Inventors: Taihei Miyamoto (Nakano, JP); Kaoru Ohashi (Sumida-ku, JP); Akira Fujiu (Mitaka, JP); Kazuki Sekiguchi (Adachi-ku, JP)

Assignee: International Business Machines Corporation

G06T11/001G06T7/90G06T2207/10024G06T2207/20081

Abstract

Claims (64)

1. A computer-implemented method for selectively replacing a color of an object in an original image, the method comprising:

identifying one or more color groups that are hard to be distinguished by a person having a color vision deficiency at a border between one or more objects in an original image;

generating a grayscale image from the original image, in response to the identification of the one or more color groups that are hard to be distinguished;

estimating an original color of each pixel in the original image by inputting the generated grayscale image to an automatic colorization artificial intelligence (AI) model, wherein the automatic colorization AI model estimates the original color of the one or more objects in the gray scale image;

determining at least one color group for which color replacement is to be performed, from the one or more color groups that are hard to be distinguished, wherein the at least one color group for which the color replacement is to be performed is a color group for which a confidence level of the estimated original color by the automatic colorization AI model is low; and

replacing the determined at least one color group with a color that is easily perceived by the person having a color vision deficiency and that is easily distinguished from the one or more color groups at the border between the one or more objects in the original image.

2. The computer-implemented method of claim 1 , further comprising:

mapping a plurality of original colors of the original image with the one or more colors that are seen by the person having the color vision deficiency;

identifying the one or more objects in the original image for which color replacement is needed, based on the mapping; and

grouping similar colors of the plurality of original colors in the identified one or more objects in the original image.

3. The computer-implemented method of claim 2 , wherein grouping similar colors of the plurality of original colors in the identified one or more objects in the original image further comprises:

grouping the colors into ranges that include a certain number of colors, wherein the ranges include several tens of colors to several hundreds of colors.

4. The computer-implemented method of claim 1 , further comprising:

replacing the determined at least one color group with a color that does not affect a perception of the one or more objects in the image.

5. The computer-implemented method of claim 1 , further comprising:

estimating a degree of strangeness feeling perceived by a viewer, based on the replaced at least one color group of the identified one or more objects in the original image; and

wherein the one or more objects that gives a strong strange feeling to the viewer, due to the replaced at least one color group, has a high confidence level in colorization by the automatic colorization AI model and is colorized with the same color as that in the original image.

6. The computer-implemented method of claim 1 , wherein the automatic image colorization AI model is a technology of learning one or more combinations of shapes and colors from a large amount of image data and colorizing a monochrome image.

7. The computer-implemented method of claim 1 , further comprising:

calculating a score for each of the adjacent objects that are hard to be distinguished; and

for each of the color groups, using a color with the lowest calculated score to replace the color group in the original image.

8. A computer program product for providing a framework to identify questions and answers dynamically from a dataset based on previous learning and an evaluation score of a user, comprising a non-transitory tangible storage device having program code embodied therewith, the program code executable by a processor of a computer to perform a method, the method comprising:

identifying one or more color groups that are hard to be distinguished by a person having a color vision deficiency at a border between one or more objects in an original image;

generating a grayscale image from the original image, in response to the identification of the one or more color groups that are hard to be distinguished;

estimating an original color of each pixel in the original image by inputting the generated grayscale image to an automatic colorization artificial intelligence (AI) model, wherein the automatic colorization AI model estimates the original color of the one or more objects in the gray scale image;

determining at least one color group for which color replacement is to be performed, from the one or more color groups that are hard to be distinguished, wherein the at least one color group for which the color replacement is to be performed is a color group for which a confidence level of the estimated original color by the automatic colorization AI model is low; and

replacing the determined at least one color group with a color that is easily perceived by the person having a color vision deficiency and that is easily distinguished from the one or more color groups at the border between the one or more objects in the original image.

9. The computer program product of claim 8 , further comprising:

mapping a plurality of original colors of the original image with the one or more colors that are seen by the person having the color vision deficiency;

identifying the one or more objects in the original image for which color replacement is needed, based on the mapping; and

grouping similar colors of the plurality of original colors in the identified one or more objects in the original image.

10. The computer program product of claim 9 , wherein grouping similar colors of the plurality of original colors in the identified one or more objects in the original image further comprises:

grouping the colors into ranges that include a certain number of colors, wherein the ranges include several tens of colors to several hundreds of colors.

11. The computer program product of claim 8 , further comprising:

replacing the determined at least one color group with a color that does not affect a perception of the one or more objects in the image.

12. The computer program product of claim 8 , further comprising:

estimating a degree of strangeness feeling perceived by a viewer, based on the replaced at least one color group of the identified one or more objects in the original image; and

wherein the one or more objects that gives a strong strange feeling to the viewer, due to the replaced at least one color group, has a high confidence level in colorization by the automatic colorization AI model and is colorized with the same color as that in the original image.

13. The computer program product of claim 8 , wherein the automatic image colorization AI model is a technology of learning combinations of shapes and colors from a large amount of image data and colorizing a monochrome image.

14. The computer program product of claim 8 , further comprising:

calculating a score for each of the adjacent objects that are hard to be distinguished; and

for each of the color groups, using a color with the lowest calculated score to replace the color group in the original image.

15. A computer system, comprising:

one or more computer devices each having one or more processors and one or more tangible storage devices; and

a program embodied on at least one of the one or more storage devices, the program having a plurality of program instructions for execution by the one or more processors, the program instructions comprising instructions for:

identifying one or more color groups that are hard to be distinguished by a person having a color vision deficiency at a border between one or more objects in an original image;

generating a grayscale image from the original image, in response to the identification of the one or more color groups that are hard to be distinguished;

estimating an original color of each pixel in the original image by inputting the generated grayscale image to an automatic colorization artificial intelligence (AI) model, wherein the automatic colorization AI model estimates the original color of the one or more objects in the gray scale image;

determining at least one color group for which color replacement is to be performed, from the one or more color groups that are hard to be distinguished, wherein the at least one color group for which the color replacement is to be performed is a color group for which a confidence level of the estimated original color by the automatic colorization AI model is low; and

replacing the determined at least one color group with a color that is easily perceived by the person having a color vision deficiency and that is easily distinguished from the one or more color groups at the border between the one or more objects in the original image.

16. The computer system of claim 15 , further comprising:

mapping a plurality of original colors of the original image with the one or more colors that are seen by the person having the color vision deficiency;

identifying the one or more objects in the original image for which color replacement is needed, based on the mapping; and

grouping similar colors of the plurality of original colors in the identified one or more objects in the original image.

17. The computer system of claim 16 , wherein grouping similar colors of the plurality of original colors in the identified one or more objects in the original image further comprises:

grouping the colors into ranges that include a certain number of colors, wherein the ranges include several tens of colors to several hundreds of colors.

18. The computer system of claim 15 , further comprising:

replacing the determined at least one color group with a color that does not affect a perception of the one or more objects in the image.

19. The computer system of claim 15 , further comprising:

estimating a degree of strangeness feeling perceived by a viewer, based on the replaced at least one color group of the identified one or more objects in the original image; and

wherein the one or more objects that gives a strong strange feeling to the viewer, due to the replaced at least one color group, has a high confidence level in colorization by the automatic colorization AI model and is colorized with the same color as that in the original image.

20. The computer program product of claim 15 , further comprising:

calculating a score for each of the adjacent objects that are hard to be distinguished; and

for each of the color groups, using a color with the lowest calculated score to replace the color group in the original image.

Assignments (1)

Continuity (1)

References Cited (16)